Research Area:  Machine Learning
The electroencephalogram (EEG) is widely used by physicians for interpretation and identification of physiological and pathological phenomena. However, the EEG signals are often corrupted by power line interferences noise and EMG induced noise. These artifacts strongly influence the utility of recorded EEGs and need to be removed for better clinical diagnosis. How to eliminate the effect of the noise is an important preprocessing problem in signal processing. In this paper, a novel and efficient power interferences reduction algorithm by the recently developed empirical mode decomposition (EMD) for the EEG signal is proposed. The principle of this method consists of decompositions of the EEG signal into a limited number of intrinsic mode function (IMF). This algorithm can effectively detect, separate and remove a wide variety of artifacts from EEG recording. Experimental results show that the proposed EMD- based algorithm is possible to achieve an excellent balance between suppresses power interference and EMG noise effectively and preserves as many target characteristics of original signal as possible.
Keywords:  
electroencephalogram
interpretation
physiological
pathological phenomena
reduction algorithm
empirical mode decomposition
Author(s) Name:  De-Xiang Zhang; Xiao-Pei Wu; Xiao-jing Guo
Journal name:  
Conferrence name:  2nd International Conference on Bioinformatics and Biomedical Engineering
Publisher name:  IEEE
DOI:  10.1109/ICBBE.2008.862
Volume Information:  
Paper Link:   https://ieeexplore.ieee.org/abstract/document/4535742